Conventional and Microwave Pyrolysis of Empty Fruit

Bunch and Rice Husk Pellets

By:

Noor Afiqah Binti Mohd

A Thesis Submitted

For the Degree of Doctor of Philosophy

Department of Chemical & Biological Engineering

University of Sheffield

February 2017

Executive Summary

i

Executive Summary In recent years, microwave pyrolysis has been the focus of intense research due to

the claim that it produced better quality products at a lower power input compared to the

electrical furnace pyrolysis system. This study aimed to investigate the influence of both

pyrolysis methods on yield and product composition obtained from Malaysian biomass,

i.e.: empty fruit bunch and rice husk pellets. They represent lignocellulosic biomass

procured as by-products of the milling process.

In the first part of the thesis, an initial characterisation of biomass was conducted

to determine the chemical composition. It was found that the biomass in this study has

moisture and volatiles content at around 5.4 wt.% and 70 wt.%, respectively which makes

them ideal for the pyrolysis process. 200g of biomass was loaded into a 15.8kW fixed-bed

pyrolysis reactor once the reactor had reached the set temperature. Typically, 40g of

biomass was pyrolysed in a specially designed 1000W multi-mode microwave oven, where

microwaves were fed into the oven cavity through a bottom-feed waveguide.

It was found that microwave pyrolysis gave a higher bio oil and char yield than

conventional pyrolysis at a similar reaction temperature. Up to 8.40% increase in bio oil

yield was observed when rice husk pellets were pyrolysed under microwave radiation at

800ºC. GC-MS analysis revealed a greater content of mono-aromatics compounds obtained

from microwave pyrolysis oils with negligible Polycyclic Aromatic Hydrocarbons (PAH)

than conventional pyrolysis oils. Similarly, greater cracking of heavier hydrocarbons at

high temperature resulted in up to 44% increase in phenol formation from microwave

pyrolysis oils.

Executive Summary

ii

A maximum surface area of 410m2/g was also recorded during microwave pyrolysis

of rice husk pellets at 500ºC, where this value reduces with an increase in pyrolysis

temperature. Moreover, microwave pyrolysis resulted in up to 29% increase in syngas

(H2+CO) evolution and about 42% lower greenhouse gases (CH4+CO2) than conventional

pyrolysis. These differences can be attributed to internal heat generation during microwave

processing in contrast to conduction from the surface inwards during conventional heating.

Energy yield analysis suggested that microwave pyrolysis can be optimised for the

production of high quality char and bio oil. Meanwhile, conventional pyrolysis can be

optimised to enhance syngas production.

The second part of this thesis looks into the effect of waveguide position and

biomass bed height on the electric field and its corresponding temperature distribution.

Numerical modelling has shown that higher temperature rise can be generated in a larger

load due to greater microwave power deposited. Moreover, an increase in relative

permittivity was observed as biomass was converted into char during pyrolysis. This

showed that microwave pyrolysis of biomass can be a self-sustaining process, without any

addition of microwave absorber.

It was concluded that viable industrial application of microwave pyrolysis is very

promising

Acknowledgements

iii

Acknowledgements

I wish to express sincere gratitude to my supervisors; Professor Jim Swithenbank,

Dr. Vida Sharifi and Dr. Grant Wilson for their endless support and supervision. Many

thanks to David Palmer, Mike O’Meara, and Oz MacFarlane for providing technical

support. Special thanks to Dr. M Sandhu from the Institute of Microwave and Photonics,

Leeds University for assistance with dielectric characterisation. Special thanks to Majlis

Amanah Rakyat (MARA) and Universiti Kuala Lumpur (UniKL) for providing financial

assistance over the course of my study. To my family and friends, thank you.

Table of Contents

iv

Table of Contents

Executive Summary .......................................................................................................................... i

Acknowledgements ......................................................................................................................... iii

Table of Contents ............................................................................................................................ iv

List of Tables ................................................................................................................................. vii

List of Figures ............................................................................................................................... viii

Nomenclatures ................................................................................................................................ xi

1 Introduction .............................................................................................................................. 1

1.1 Current World Energy, Environment and Economy Scenario ......................................... 1

1.2 Greenhouse Gas Emission ............................................................................................... 2

1.3 Climate Change and Energy Security .............................................................................. 4

1.4 Emissions Target .............................................................................................................. 5

1.5 Costs of Technologies ...................................................................................................... 6

1.6 Biomass ............................................................................................................................ 8

1.7 Biomass Conversion Technologies .................................................................................. 9

1.8 Aim and Objectives of the Present Work ....................................................................... 11

1.9 Novelty ........................................................................................................................... 12

1.10 Thesis Layout ................................................................................................................. 12

2 Literature Review ................................................................................................................... 13

2.1 Introduction to Biomass ................................................................................................. 13

2.1.1 Lignocellulosic Biomass ............................................................................................ 14

2.2 Biomass Properties ......................................................................................................... 16

2.3 Pyrolysis ......................................................................................................................... 20

2.3.1 Mechanism ................................................................................................................. 20

2.3.2 Cellulose Decomposition ........................................................................................... 22

2.3.3 Bio Oil Properties....................................................................................................... 25

2.3.4 Tar Composition ........................................................................................................ 26

2.3.5 Pyrolysis Types .......................................................................................................... 28

2.4 Microwave Heating ........................................................................................................ 28

2.4.1 Microwave Modes...................................................................................................... 31

2.4.2 Conventional Heating versus Microwave Heating .................................................... 32

2.5 Previous Work on Microwave Pyrolysis of Biomass .................................................... 33

2.6 Review on Numerical Simulation .................................................................................. 38

2.6.1 Microwave Heating .................................................................................................... 38

Table of Contents

v

2.6.2 Transient Heating ....................................................................................................... 40

2.7 Summary ........................................................................................................................ 41

3 Materials and Methods ........................................................................................................... 42

3.1 Biomass Feedstock ......................................................................................................... 42

3.2 Conventional Pyrolysis ................................................................................................... 44

3.2.1 Overview .................................................................................................................... 44

3.2.2 Methodologies ............................................................................................................ 47

3.3 Microwave Pyrolysis ...................................................................................................... 48

3.3.1 Overview .................................................................................................................... 48

3.3.2 Methodologies ............................................................................................................ 52

3.3.3 Calculation of Pyrolysis Product Yield ...................................................................... 53

3.3.4 Comparison between Fixed and Variable Power Configuration ................................ 53

3.4 Description of Temperature Measurement System ........................................................ 55

3.4.1 Microwave Thermocouple Effect ............................................................................... 56

3.5 Solid Analysis ................................................................................................................. 58

3.5.1 Ultimate Analysis ....................................................................................................... 58

3.5.2 Trace Analysis ............................................................................................................ 59

3.5.3 Proximate Analysis ..................................................................................................... 60

3.5.4 Calorific Value ........................................................................................................... 61

3.5.5 Scanning Electron Microscopy ................................................................................... 62

3.5.6 B.E.T Surface Area ..................................................................................................... 63

3.5.7 Dielectric Properties ................................................................................................... 64

3.6 Liquid Analysis .............................................................................................................. 66

3.6.1 Ultimate Analysis ....................................................................................................... 66

3.6.2 FTIR Analysis ............................................................................................................ 66

3.6.3 GC-MS Analysis ........................................................................................................ 67

3.7 Gas Analysis ................................................................................................................... 68

3.8 Error Analysis ................................................................................................................. 69

3.9 Summary ........................................................................................................................ 70

4 Results and Discussion of Conventional and Microwave Pyrolysis ...................................... 71

4.1 Characterisation of Biomass Feedstock in This Study ................................................... 71

4.1.1 Analysis of Trace Elements Present in Biomass ........................................................ 73

4.1.2 Derivative Thermogravimetry Analysis ..................................................................... 74

4.2 The Effect of Temperature and Heating Process on Pyrolysis Product Yield ................ 76

4.3 Char Analysis ................................................................................................................. 80

4.3.1 Ultimate and Proximate Analysis ............................................................................... 80

4.3.2 Calorific Value ........................................................................................................... 82

4.3.3 Scanning Electron Microscopy ................................................................................... 83

4.3.4 B.E.T Specific Surface Area....................................................................................... 88

4.3.5 Relative Permittivity ................................................................................................... 89

4.4 Bio Oil Analysis ............................................................................................................. 90

Table of Contents

vi

4.4.1 Ultimate Analysis ....................................................................................................... 90

4.4.2 GC-MS Analysis ........................................................................................................ 92

4.4.3 FT-IR Analysis ........................................................................................................... 96

4.5 Gas Analysis .................................................................................................................. 99

4.6 Energy Consumption ................................................................................................... 103

4.6.1 Energy Balance Analysis ......................................................................................... 106

4.6.2 Energy Yield ............................................................................................................ 109

4.7 Summary ...................................................................................................................... 112

5 Numerical Simulation of Microwave & Conventional Heating........................................... 114

5.1 Microwave Heating Simulation ................................................................................... 114

5.1.1 Governing Equations................................................................................................ 114

5.1.2 Geometric Model ..................................................................................................... 115

5.1.3 Boundary Conditions ............................................................................................... 120

5.2 Results and Discussion of Microwave Heating Simulation ......................................... 122

5.2.1 The Effect of Input Properties .................................................................................. 122

5.2.2 The Effect of Load Size ........................................................................................... 123

5.2.3 The Effect of Waveguide Location .......................................................................... 128

5.2.4 Modelling Microwave Thermocouple Effect ........................................................... 131

5.3 Conventional Heating Simulation ................................................................................ 134

5.3.1 Governing Equations................................................................................................ 135

5.4 Results and Discussion of Conventional Heating Simulation ...................................... 137

5.4.1 Effect of Particle Size .............................................................................................. 137

5.5 Summary ...................................................................................................................... 139

6 Challenges and Opportunities of Pyrolysis Process ............................................................. 141

6.1 Issues Associated with the Scale-Up of Microwave Pyrolysis .................................... 141

6.2 Market for Pyrolysis Products...................................................................................... 142

6.3 Potential for Biomass Pyrolysis ................................................................................... 144

7 Conclusions and Suggestions for Future Work .................................................................... 146

7.1 Conclusions ........................................................................................................................ 146

7.2 Suggestions for Future Work ....................................................................................... 148

8 References ............................................................................................................................ 149

List of Tables

vii

List of Tables

Table 1-1: Levelised cost of electricity for projects starting 2018, at 10% discount rate

(Department of Energy & Climate Change, 2012) 7

Table 2-1: Biomass major groups. Adapted with permission from (Basu, 2010b) 13

Table 2-2: Biomass properties. Published with permission from (Jenkins et al., 1998) 17

Table 2-3: List of tar compounds. Published with permission from (Li & Suzuki, 2009) 27

Table 2-4: Types of pyrolysis. Adapted from (Jahirul et al., 2012) 28

Table 4-1: Properties of biomass feedstock 72

Table 4-2: Minor elements of biomass feedstock 74

Table 4-3: Chemical properties of EFB chars relative to raw biomass 80

Table 4-4: Chemical properties of rice husk chars relative to raw biomass 81

Table 4-5: Dielectric properties of biomass samples at f=2.45 GHz and T=298K 89

Table 4-6: Chemical properties of EFB bio oils 91

Table 4-7: Chemical properties of rice husk bio oils 92

Table 4-8: GC-MS analysis of EFB oils (μg/L) 93

Table 4-9: GC-MS analysis of rice husk oils (μg/L) 94

Table 4-10: Chemical compounds in biomass pyrolysis oils 98

Table 4-11: Energy content of pyrolysis gas 100

Table 4-12: Energy consumption during conventional and microwave pyrolysis of EFB pellets 105

Table 4-13: Energy consumption during conventional and microwave pyrolysis of rice husk

pellets 105

Table 4-14: Energy balance of EFB pyrolysis 108

Table 4-15: Energy balance of rice husk pyrolysis 108

Table 5-1: Material properties used in this study 119

Table 5-2: Domain mesh defined in this model 119

Table 5-3: Input properties used during simulation 135

Table 5-4: Time taken for centre of particle to reach equilibrium with T∞ 139

List of Figures

viii

List of Figures Figure 1-1: Energy consumption by sector in 2015 (Doman, 2016) 2

Figure 1-2: CO2 emission (The World Bank, 2016) 3

Figure 1-3: Crude oil prices (U.S Energy Information Administration, 2016) 5

Figure 1-4: Biomass conversion technologies. Adapted from (Basu, 2010a) 10

Figure 2-1: Schematic of lignocellulosic components of biomass. Published with permission from

(Yin, 2012) 15

Figure 2-2: Bulk density of biomass relative to coal (Clarke & Preto, 2011) 19

Figure 2-3: Pyrolysis stages. Published with permission from (Neves et al., 2011) 21

Figure 2-4: Broido model of cellulose decomposition (Varhegyi & Jakab, 1994) 23

Figure 2-5: Broido-Shafizadeh model of cellulose decomposition. Published with permission from

(Basu, 2010c) 24

Figure 2-6: Tar formation scheme. Published with permission from (Li & Suzuki, 2009) 27

Figure 2-7: Electromagnetic spectrum. Published with permission from (Motasemi & Afzal,

2013) 29

Figure 2-8: Electric (E) and magnetic (H) field components in microwave. Published with

permission from (Motasemi & Afzal, 2013) 30

Figure 2-9: Complex permittivity of dielectric material 31

Figure 2-10: Difference in heating process. Published with permission from (Motasemi & Afzal,

2013) 33

Figure 2-11: Cross section of wood block, ϕ= 80 mm. Published with permission from (Miura et

al., 2004) 35

Figure 2-12: Coupling of electromagnetic field and temperature distribution 39

Figure 2-13: Radial conduction in a cylindrical biomass particle 40

Figure 3-1: Empty fruit bunch (QM Consultants, 2016) 43

Figure 3-2: Rice husk (Stylus, 2016) 43

Figure 3-3: Conventional pyrolysis setup 45

Figure 3-4: Front view of pyrolyser 45

Figure 3-5: Product recovery setup 46

Figure 3-6: Tar clean-up 46

Figure 3-7: Microwave pyrolysis setup 48

Figure 3-8: Front view of the BP-125 microwave oven 49

Figure 3-9: Volatiles extraction and temperature measurement setup 50

Figure 3-10: Data acquisition setup 50

Figure 3-11: Reaction vessel in a refractory furnace 51

List of Figures

ix

Figure 3-12: Mode stirrer 51

Figure 3-13: Temperature profiles during MP EFB at 500°C 56

Figure 3-14: Temperature profile during MP EFB at 800°C 56

Figure 3-15: Thermocouple reading during microwave heating and natural cooling 57

Figure 3-16: Time constant (τ) of thermocouple made with ungrounded junction type

(Engineering, 2017) 58

Figure 3-17: Thermo Scientific Flash 2000 Organic Elemental Analyser 59

Figure 3-18: Thermogravimetry analyser 60

Figure 3-19: Parr 6200 calorimeter 61

Figure 3-20: Philips XL30S FEG 63

Figure 3-21: Surface characterisation analyser 64

Figure 3-22: Vector Network Analyser connected to an open ended micro-strip stub 65

Figure 3-23: Perkin Elmer Frontier spectrometer 67

Figure 3-24: Shimadzu GCMS-QP2010 68

Figure 3-25: Thermo Scientific Trace 1310 Gas Chromatograph 69

Figure 4-1: Biomass pellets used in this study. A) EFB pellets and B) Rice husk pellets 72

Figure 4-2: Major elements of biomass feedstock 74

Figure 4-3: DTG curves 75

Figure 4-4: Product yield from EFB pyrolysis 76

Figure 4-5: Product yield from rice husk pyrolysis 77

Figure 4-6: Calorific value of chars relative to raw biomass 83

Figure 4-7: (L) MP RH char at 800°C and (R) Small globules on CP RH at 800°C 84

Figure 4-8: SEM images of rice husk char obtained from top: conventional pyrolysis, bottom:

microwave pyrolysis at 500°C 85

Figure 4-9: SEM images of MP EFB char at 500°C 86

Figure 4-10: (L) MP RH char at 800°C, and (R) MP EFB char at 800°C 87

Figure 4-11: Energy Dispersive Spectroscopy (EDS) spectrum of MP RH char at 800°C 88

Figure 4-12: Specific surface area of pyrolysed rice husk char 89

Figure 4-13: FTIR spectra of EFB oils at 500°C 96

Figure 4-14: FTIR spectra of EFB oils at 800°C 97

Figure 4-15: FTIR spectra of rice husk oils at 500°C 97

Figure 4-16: FTIR spectra of rice husk oils at 800°C 98

Figure 4-17: Gas evolution during EFB pyrolysis (N2 free) 99

Figure 4-18: Gas evolution during rice husk pyrolysis (N2 free) 100

Figure 4-19: Power input during conventional pyrolysis 103

List of Figures

x

Figure 4-20: Power input during microwave pyrolysis 104

Figure 4-21: Pyrolysis system control volume 107

Figure 4-22: Energy yield from conventional pyrolysis of EFB pellets 110

Figure 4-23: Energy yield from microwave pyrolysis of EFB pellets 110

Figure 4-24: Energy yield from conventional pyrolysis of rice husk pellets 111

Figure 4-25: Energy yield from microwave pyrolysis of rice husk pellets 111

Figure 5-1: Geometric model of microwave oven 116

Figure 5-2: Biomass bed domain 117

Figure 5-3: Biomass bed domain in a Cartesian plane. (Illustration not drawn to scale) 118

Figure 5-4: Symmetry boundary 120

Figure 5-5: Port boundary 121

Figure 5-6: Impedance boundary 121

Figure 5-7: Influence of thermal properties on the rate of heating. Bed height, h=50mm 122

Figure 5-8: Electric field distribution in an unloaded cavity at f= 2.45 GHz 123

Figure 5-9: Electric field distribution in a loaded cavity at f=2.45 GHz. Load size: (ϕ 29 x 40mm)

124

Figure 5-10: Standing wave pattern and temperature distribution of biomass at a different bed

height 126

Figure 5-11: Temperature profile across biomass bed at different radiation time.

Load size: (ϕ= 29 x 50 mm). 127

Figure 5-12: Waveguide position in a) Daewoo 500 W microwave oven, b) microwave oven in

this study 128

Figure 5-13: Bottom-side waveguide 129

Figure 5-14: Right-side waveguide 129

Figure 5-15: Top-side waveguide 129

Figure 5-16: Microwave power absorbed at a different bed height 131

Figure 5-17: Geometric model 132

Figure 5-18: Temperature distribution inside biomass bed without a thermocouple 132

Figure 5-19: Temperature distribution inside biomass bed with thermocouple 133

Figure 5-20: Magnified view 133

Figure 5-21: Temperature profile at Bi= 0.64 138

Figure 5-22: Temperature profile at Bi= 1 138

Nomenclatures

xi

Nomenclatures Abbreviations

BET Brunauer-Emmett-Teller

CCGT Combined Cycle Gas Turbine

CHP Combined Heat and Power

CP Conventional Pyrolysis

DTG Derivative Thermogravimetry

EDS Energy Dispersive Spectroscopy

EFB Empty Fruit Bunch

EM Electromagnetic

FIT Feed-In Tariff

FTIR Fourier Transform Infrared

GCV Gross Calorific Value

GHG Greenhouse Gases

LPM Litre Per Minute

MP Microwave Pyrolysis

PAH Polycyclic Aromatic Hydrocarbons

RH Rice Husk

SEM Scanning Electron Microscopy

SVOC Semi-Volatile Organic Compounds

TGA Thermogravimetry Analysis

VOC Volatile Organic Compounds

Symbol Description SI unit

∆T Temperature difference ºC

B Magnetic flux density T

Bi Biot number -

Cp Heat capacity at constant pressure J/kgK

E Electric field intensity V/m

e1 Correction for heat of formation HNO3 J

e2 Correction for heat of formation of H2SO4 J

e3 Correction for heat of combustion of fuse wire J

Eb Energy content of biomass J/kg

Ei Energy content of pyrolysis product J/kg

Nomenclatures

xii

Es Source electric field V/m

f Frequency Hz

h Convective heat transfer coefficient W/m2K

H Magnetic field intensity A/m

J Current density A/m2

k Thermal conductivity W/mK

k0 Wave number of free space rad/m

l Length m

Lc Characteristic length m

m Mass kg

N Complex refractive index -

P Partial pressure of N2 Pa

P0 Saturation pressure of N2 Pa

Qb Energy content of biomass J/kg

Qc Energy content of pyrolysed char J/kg

Qe Electrical energy J/kg

Qg Energy content of pyrolysis gas J/kg

Qloss Energy loss within the system J/kg

Qoil Energy content of pyrolysis bio oil J/kg

r Radial position m

T Temperature K

t Time s

T0 Initial temperature of biomass K

T∞ Bulk temperature K

tan α Loss factor -

v Volume of adsorbed gas L

vm Volume of adsorbed gas in a monolayer L

W Energy Equivalent J/ºC

x Dimensionless distance -

Yi Product yield -

β Propagation constant rad/m

ΔY Mass loss kg

ϵ Emissivity -

ε Permittivity F/m

ϵ’ Dielectric constant -

Nomenclatures

xiii

ϵ’’ Dielectric loss -

ε0 Permittivity of free space (8.85 x 10-12) F/m

εr Relative permittivity -

η Efficiency -

θ Dimensionless temperature -

μ Permeability H/m

μr Relative permeability -

ρ Density m3/kg

σ Stefan-Boltzmann constant (5.67 x 10-8) W/m2K4

τ Dimensionless time -

ϕ Diameter m

ω Angular frequency rad/s

𝛼 Thermal diffusivity m2/s

Chemical Compounds

CH4 Methane

CO Carbon monoxide

CO2 Carbon dioxide

H2 Hydrogen

H2O Water

K Potassium

N2 Nitrogen

Si Silica

SiO2 Silicon Dioxide

Introduction

1

1 Introduction

This chapter looks into the current world energy scenario, the accumulation of waste

biomass worldwide, and the cost of conversion technologies. It also looks into global

warming as an imminent threat to humanity and pyrolysis as a solution to waste

management and energy recovery. The aim and objectives of the present work are

highlighted, along with thesis outline at the end of this section.

1.1 Current World Energy, Environment and Economy Scenario

The world’s energy consumption is projected to increase by 48% between 2012 and

2040 with most demand coming from developing countries outside the Organisation for

Economic Cooperation and Development (Doman, 2016). Energy demand is projected to

reach above 8.44 x 1014 MJ. Emerging markets are projected to grow by 4.7% in 2017, with

India as the fastest growing economy at a rate of 7.5%. India for instance, is the world’s

third largest producer of crude steel at 88.98 MT. This market is expected to grow further

driven by growth in infrastructure development and rising demand from automotive

industry (India Brand Equity Foundation, 2016). Steel manufacturing is one of the most

energy-intensive industries. With this in mind, the nation is required to secure a stable

supply of energy to sustain its long-term economic development.

Moreover, global population is projected to hit 9.7 billion mark by 2050, a 9%

increase from 6.1 billion in 2000. This represents a serious need to address the solution in

terms of food and energy security, resource and waste management to create a sustainable

development without compromising the environment. Increasing energy demand would put

a strain on the old centralised power generation and distribution.

Introduction

2

Figure 1-1: Energy consumption by sector in 2015 (Doman, 2016)

Primary energy supplies include coal, oil, and natural gas, along with smaller share

of renewables. The breakdown of primary energy consumption is illustrated in Figure 1-1.

According to the data released by U.S Energy Information and Administration (Doman,

2016), almost 40% or 4.02 x 1013 MJ of energy supplies goes into electricity generation.

This is followed by transportation sector at 28.33%, fuelled mostly by oil and liquefied

natural gas (LNG). The combination of the industrial, residential, and commercial sectors

takes up around 32.60% of total energy supplies. Energy consumption and greenhouse gas

(GHG) emissions are closely related. The former is driven by factors such as energy

demand, climatic conditions, combustion engine inefficiency, and poor building insulation.

For example, high energy consumption is expected during cold winter months for

commercial and residential heating.

1.2 Greenhouse Gas Emission

Primary constituents of GHG include carbon dioxide, methane, nitrous oxide,

chlorofluorocarbons, and hydro chlorofluorocarbons (known collectively as Freon gases).

These gases are present in the lower atmosphere, with capability to absorb some of the

6.67%4.17%

21.76%

28.33%

39.08%

Residential Commercial Industrial Transport Electricity generation

Introduction

3

outgoing radiation leaving the earth. This in turn helps to keep the earth warm. However,

activities such as deforestation, the burning of fossil fuels, and livestock farming are mainly

responsible for anthropogenic GHG emissions. Since the beginning of Industrial

Revolution, heavy emission of GHG into the atmosphere has altered the equilibrium of the

natural carbon cycle.

Figure 1-2: CO2 emission (The World Bank, 2016)

Heavy reliance on traditional fossil fuels to fuel our economic growth has its own

set of challenges. Total GHG emission in 2012 stands at 52.8 Giga Tonne (GT) of

CO2 equivalent. Carbon dioxide made up the largest fraction of total emissions at 34.65 GT

in 2011, rising by 268.7% from the 1960 levels, as shown in Figure 1-2. Fossil fuel

combustion, cement and steel manufacturing, and gas flaring are some of the main

contributors for increased CO2. Meanwhile, the concentration of methane in the

atmosphere is alarming due to its high global warming potential (GWP). It is a measure of

energy absorption of gas emitted, relative to carbon dioxide. In this case, methane is 25

times better at retaining heat than carbon dioxide, within 100 year time frame (United

Nations Framework Convention on Climate Change, 2014a). Intensive livestock farming

5

10

15

20

25

30

35

1960 1970 1980 1990 2000 2010

CO

2em

issi

on

(k

t)M

illi

ons

Year

Introduction

4

and poor waste management have resulted in increase of CH4 concentration. For instance,

anaerobic decomposition of waste in landfill produced biogas, which consists around 50-

55 vol. % CH4 and 45-50 vol. % CO2. However, the release of these gases could be avoided

by utilising a proper landfill design with biogas collection. Other impacts of GHG

emissions include acid rain, ozone depletion, and the creation of low level ozone.

1.3 Climate Change and Energy Security

Global warming has resulted in the melting of ice caps, rising sea levels, and an

erratic global weather pattern. This is indicated by severe drought and increased in

hurricane intensity in different parts of the world, which has led to damage incurred by real

estate and income losses. Imminent effects of climate change include low crop yield and

rising food prices, where poor countries are the most vulnerable. Sir Nicholas Stern in his

Stern Review has concluded that the “benefits of early action to mitigate climate change

far outweigh the economic costs of not acting”. Delaying the actions to mitigate climate

change would cost the world 5% of its annual Gross Domestic Product (GDP), due to

climate-related damage. This report strongly recommends that the global GHG levels to be

stabilised at around 450-550 ppm CO2e in order to avoid 2°C increase in global surface

temperature and mitigate the worst impact of climate change. This would require collective

effort to reduce GHG emissions by at least 25% or more below current levels by 2050

(Stern, 2006).

Introduction

5

Figure 1-3: Crude oil prices (U.S Energy Information Administration, 2016)

Commodities such as oil and natural gas are traded on the global market. It is

subjected to volatility in prices driven by factors such as geopolitical events, spare capacity,

demand and supply, and market uncertainty. Figure 1-3 shows the graph of crude oil prices

over the past 45 years. 1973 oil crisis marked the first key event that drove the crude oil

price from $15 to $42 (in 2010 USD), a year later. The Iran-Iraq war, global financial

collapse, and production cut from OPEC are among key events that influence the global

crude oil prices. This prompts the need to secure stable supply of fuels over the next

decades.

1.4 Emissions Target

The Kyoto Protocol was adopted in 1997 in the first international joint effort to curb

GHG emissions. It is based on a premise that industrialised nations were responsible for

GHG released over the past 150 years of economic activity. It bound the 37 industrialised

nations to reduce emissions by an average of 5%, from the 1990 level during its first

commitment period (2008-2012). This is achieved through Joint Implementation, Clean

Development Mechanism, and Emission Trading between developed and developing

0

25

50

75

100

125

150

1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Pri

ce p

er b

arr

el (

Rea

l U

S d

oll

ar)

Year

WTI real

Importer real partial

Introduction

6

countries. Kyoto Protocol is superseded by the Doha Amendment in 2012 for its second

commitment period between 2013 and 2020. This amendment is ratified by 66 countries,

with binding commitment to reduce GHG emission by 18%, relative to 1990 levels (United

Nations Framework Convention on Climate Change, 2014b).

The emissions target, resource depletion, energy security, population growth, and

climate change issues have triggered the need to find alternative fuels from renewable

energy (RE) to ensure sustainable development. The current fossil-fuel based energy

systems are not only environmentally unsustainable, they are also highly inequitable,

leaving 1.4 billion people without access to electricity (United Nations Environment

Programme, 2016). Renewable energy differs from non-renewables in the way that it

replenishes faster for efficient extraction. Sources of renewable energy include wind, tidal,

solar, geothermal, hydro, and biomass. With most emissions coming from the combustion

of fossil fuels in power and transport sector, it is vital to switch into the low carbon

technologies.

1.5 Costs of Technologies

High generation cost is one of the biggest challenges in the implementation of low

carbon technologies. It is reflected by high electricity price which is passed on to the end

consumer. Levelised cost of electricity (LCOE) is an average cost over the lifetime of

generation plant, converted into equivalent unit of cost of generation in (£/MWh). It takes

into consideration costs incurred during planning, construction, operating and

decommissioning stages. It covers the capital cost, fixed and variable operational cost, fuel

cost, and carbon tax. It provides assessment for different conversion technologies without

Introduction

7

taking into account revenue streams generated; e.g.: sale of electricity. Cost estimates for a

project starting in 2018 are presented in Table 1-1.

Table 1-1: Levelised cost of electricity for projects starting 2018, at 10% discount rate

(Department of Energy & Climate Change, 2012)

Type Technologies LCOE (£/MWh)

Gas CCGT 85

CCGT with post comb. CCS 94

Biomass

< 50MW 115

>50MW 121

Coal

ASC with FGD 113

ASC with post comb. CCS 116

IGCC 131

IGCC with CCS 111

Nuclear 73

Wind

Onshore >5MW, E&W 101

Onshore >5MW, UK 90

Offshore R2 103

Offshore R3 113

Gas and coal-fired power plants are equipped with post-combustion technologies

such as Flue Gas Desulphurisation (FGD) and Carbon Capture and Storage (CCS), in an

effort to reduce emission of GHG and pollutants, e.g.: SO2 into the atmosphere.

Development in high-temperature steel that is capable to withstand super-critical steam has

seen the progress of Advanced Super-Critical Coal (ASC) which aims to increase net

electrical efficiency. Meanwhile, in the Integrated Gasification Combined Cycle (IGCC),

coal is first gasified to produce synthesis gas and cleaned to remove impurities prior to

combustion in a gas turbine. These relatively new technologies together with carbon tax

are making power generation from coal expensive. For example, ASC with post

combustion CCS would cost £116/MWh.

Introduction

8

Nuclear power generation although financially attractive due to cheap fuel cost at

£73/MWh is faced with uncertainties over public perception fuelled by recent nuclear

incidents. However, biomass conversion technologies are not particularly inexpensive,

either. Large generation plant (>50 MW) would cost £121/MWh. Fuel cost contributes a

large fraction of total cost at 53.72%. This can be attributed to demanding fuel handling

requirement, fuel transport, and storage facilities. This includes pre-treatment processes;

e.g.: grinding and drying and post-combustion clean-up. Rising fossil-fuel prices and

carbon tax levied on fossil fuel generators could drive biomass to be cost competitive in

the generation market.

1.6 Biomass

Biomass is a precursor of coal formation, described as organic matter of recent

biological origin. Incorporating biomass into the energy generation mix provides several

advantages. The global potential of biomass is very broad, estimated at around

33-1135 EJ/year within 50 years timeframe, with agricultural and forestry residues

accounting for most of these wastes (Hoogwijk et al., 2003). Daioglou et al. (Daioglou et

al., 2016) put this figure at 120 EJ/year, and theoretically it could increase up to

140-170 EJ/year. Two-thirds of these comes from the high residue yield plantations in Asia

and North America regions. It provides energy security and sustainable development as

biomass energy generation does not contribute to natural carbon source; i.e.: CO2 neutral.

This is based on a premise that carbon dioxide absorbed by a plant during its lifetime is

released to the atmosphere during energy conversion. This could reduce the global carbon

footprint in an effort to mitigate global warming.

Introduction

9

Biomass has traditionally been used in low energy applications such as domestic

cooking and heating in a rural community. Biomass such as wood, charcoal and animal

manure is combusted indoor with low conversion efficiency, leading to soot emission

which imposes a respiratory health risk. Globally, agricultural residues have been used as

feed in biomass-fired CHP plants across Scandinavia, Latin America, and India. For

instance, Brazil as the major sugarcane producer has been burning sugarcane bagasse in its

co-generation plants, with surplus electricity exported to the utility grid. Sweden has set

the best example with its commitment towards renewable energy generation. Currently,

66.4% or 246 TWh of total domestic energy in Sweden is supplied by a mix of renewable

sources. The total share of bioenergy generation has surpassed oil at 35.2%, with forestry

residues as the primary source of fuel (The Swedish Bioenergy Association, 2016).

1.7 Biomass Conversion Technologies

Biomass conversion technologies exploit the stored chemical energy within a

biomass structure into another form of useful energy. Figure 1-4 highlights the major routes

for energy conversion from biomass, which includes biochemical and thermochemical

technologies. Choice of conversion technology is heavily influenced by feedstock

availability, desired end user, environmental standard, and total cost (Saidur et al., 2011).

Biochemical conversion is ideal for ‘wet’ biomass with more than 50% water content,

e.g.: sewage sludge. Some of the techniques include anaerobic digestion, fermentation, and

esterification.

Introduction

10

Figure 1-4: Biomass conversion technologies. Adapted from (Basu, 2010a)

Thermal conversion technologies include combustion, gasification, and pyrolysis.

Direct combustion or incineration is considered a mature technology, while gasification

and pyrolysis are still less proven at industrial scale. The main difference between thermal

conversion technologies is the oxygen supplied during the process. Combustion under

stoichiometric condition generates heat that can be used in a steam or gas turbine. Partial

oxidation of fuel or gasification is designed to optimise the production of syngas, e.g.: CO,

H2, and CH4. These high calorific value (CV) gases can be used directly in internal

combustion engines or synthesised into liquid fuels under Fischer-Tropsch process.

Meanwhile, pyrolysis is thermal decomposition of biomass in an inert environment.

Unlike combustion process, pyrolysis has garnered attention due to the prospect of valuable

product recovery, ranging from char and bio oil; to gas. For example, lignocellulosic phenol

in bio oil could be a substitute for petroleum-based phenol. The release of moisture and

volatiles during pyrolysis leaves behind carbon-rich char of high calorific value. In short,

the pyrolysis process has the ability to yield a range of value-added products relative to the

initial feedstock.

Introduction

11

Until recently, the pyrolysis process has been conducted in an externally-heated

reactor where the mode of heat transfer is by conduction, convection, and radiation (Chen

et al., 2008). This process is often subjected to several limiting factors which influence

product yield and distribution. The current study is looking to exploit microwave radiation

to conduct the pyrolysis process, with the aim to produce higher yield and better quality

products than conventional pyrolysis.

1.8 Aim and Objectives of the Present Work To the best of this author’s knowledge, no comparative study has been done so far

on the thermal behaviour of empty fruit bunch and rice husk pellets under microwave

heating and comparing it with conventional heating. This study aims to investigate the

effect of the heating process and temperature on the yield and characteristics of char, bio

oil and syngas. The experiments were carried out using a fixed-bed pyrolysis reactor and a

modified laboratory microwave oven at varying sample temperature, by keeping other

parameters constant. Characterisation of biomass feedstock before and after pyrolysis

process was also performed. This covers various analyses such as proximate analysis,

ultimate analysis, dielectric properties, heating value, FT-IR, GC-MS, surface area analysis

and surface image analysis.

Moreover, the detailed behaviour of the electromagnetic field and the effect on

dielectric heating of biomass in a multi-mode microwave oven is not fully understood. A

multi-physics software package was used to model the microwave heating process. This

fundamentally based study is also designed to investigate the effect of waveguide location

and biomass bed height on microwave heating efficiency. This is compared with transient

heat conduction in conventional heating. Finally, this original study seeks to analyse the

Introduction

12

savings associated with the microwave pyrolysis process for the selected Malaysian

materials and assesses the potential for industrial scale up.

1.9 Novelty Comparative study on the characterisation of liquids obtained from conventional

and microwave pyrolysis can be considered as novelty of the present work. To the best of

author’s knowledge, no such study has been done on pyrolysis bio oil. It has successfully

demonstrated the effect of microwave technology in eliminating the long-chain

hydrocarbons and increasing the presence of mono-aromatics in bio oil.

1.10 Thesis Layout

Chapter 1 explores the underlying issues that rationalise pyrolysis as a route to

energy recovery. Chapter 2 looks into the literature and previous work done in this area,

including the review of numerical simulation of microwave and conventional heating.

Meanwhile, chapter 3 provides a detailed description of experimental setups and

methodologies used for biomass and pyrolysis product characterisation. Chapter 4

discusses the properties of biomass materials used in this study and the influence of

temperature and heating method on pyrolysis product distributions. This section also sets

procedures to determine the efficiency of a pyrolysis system. This is followed by numerical

simulation of microwave and transient heating in Chapter 5. Meanwhile, Chapter 6 looks

into the challenges and opportunities associated with the scale-up of microwave processing

of biomass for biofuels production. Finally, conclusions derived from this study along with

some suggestions for future work are highlighted in Chapter 7.

Literature Review

13

2 Literature Review

This chapter provides a critical review of the work that has been done in this area.

It starts with providing an overview of biomass composition, pyrolysis mechanism and the

properties of pyrolysis products. The difference between conventional and microwave

pyrolysis is also discussed. A summary of previous work done in the area of microwave

processing of biomass is presented, followed by a section on numerical simulation of

microwave and transient heating.

2.1 Introduction to Biomass

Different interpretation exists with regards to biomass classification. Biomass can

be grouped into two major categories; virgin and waste biomass. This is highlighted in

Table 2-1. In essence, virgin biomass is extracted from primary sources and is grown for

specific purpose, i.e.: energy crops. Meanwhile, waste biomass is a secondary source of

biomass-derived products such as municipal solid waste, agricultural residues, and cooking

oil. They represent a prospect for valuable energy recovery, turning waste into useful

products.

Table 2-1: Biomass major groups. Adapted with permission from (Basu, 2010b)

Biomass type Sub-class Examples

Virgin Terrestrial biomass Grasses, energy crops, cultivated crops

Aquatic biomass Algae, water plant

Waste

Municipal waste Municipal solid waste (MSW), sewage sludge, landfill gas

Agricultural solid waste Livestock manure, agricultural crop residue

Forestry residues Bark, leaves, floor residues

Industrial wastes Demolition wood, sawdust, waste cooking oil

Literature Review

14

2.1.1 Lignocellulosic Biomass

The photosynthesis process utilises water and carbon dioxide in the presence of

light to produce carbohydrate and oxygen, which are essential for growth as outlined in

Eq: 2-1 (Basu, 2010b) . Some parts of plants; e.g.: stem, straw, husk are fibrous and are

deemed unfit for human consumption. These components are known as lignocellulose.

Unlike sugar-rich crops, e.g.: corn that is digestible through fermentation, lignocellulosic

biomass requires a different treatment.

6𝐻2𝑂 + 6𝐶𝑂2 → 6𝐶𝐻2𝑂 + 6𝑂2 Eq: 2-1

Lignocellulosic biomass is made up of hemicellulose, cellulose, lignin, and a

smaller number of extractives. A schematic of lignocellulosic components of biomass is

illustrated in Figure 2-1. According to Vassilev et al. (Vassilev et al., 2010), their

composition varies between biomass depending on biomass type, species, growth process,

and growing condition. Hemicellulose (C5H8O4)n is a short branched polymer made up of

five to six sugar monomers. Glucose, xylose, mannose, galactose, and arabinose constitute

the hemicellulose. It has an amorphous structure with low degree of polymerisation

(~100-200), making it suitable for hydrolysis. It thermally decomposes at low temperature

at around 200-260°C, and is mainly responsible for the production of non-condensable

gases and less tar.

Literature Review

15

Figure 2-1: Schematic of lignocellulosic components of biomass. Published with permission from

(Yin, 2012)

Unlike hemicellulose, cellulose (C6H10O5)n is a long chain polymer made up of

glucose monomers. It makes up the major components of lignocellulosic biomass, at around

40 to 80 wt.%. Its crystalline structure with high degree of polymerisation ~10,000, gives

it a better thermal degradation than hemicellulose. Cellulose thermally degrades at a

temperature around 240-350°C (Yin, 2012) .

Meanwhile, lignin is the primary building block of plant cell wall which provides

structural strength and resistance against microbial infection. It is made up of branched

polymers of phenyl-propanoid, or 4-propenyl phenol, 4-propenyl-2-methoxy phenol, and

4-propenyl-2.5-dimethoxyl phenol, held together by aromatic benzene rings (Diebold,

1994). The presence of ether bonds and carbon-carbon bonds give it a low degree of

oxidation and thus, a higher heating value. Lignin decomposes at a temperature higher than

that of cellulose and hemicellulose, at around 280-500°C (Mohan et al., 2006). It is mainly

responsible for the formation of char and phenol found in bio oil. However, the presence

of lignin in plant cell wall hinders enzymatic hydrolysis of carbohydrates, i.e.: cellulose

from taking place. This is due to high insolubility of lignin in acid. For this reason, thermal

Literature Review

16

conversion of lignocellulosic biomass is preferable to biochemical process. The

biochemical conversion route would require pre-treatment process to break the lignin wall

and loosening the crystalline structure of cellulose and hemicellulose.

An example of pre-treatment is steam explosion, which is aimed to destroy the fibril

structure of plant cell wall. During this stage, biomass is subjected to high pressure steam

during a short period of time, and depressurised back to atmospheric pressure. This is a

complex process where considerable amount of by-products must be removed before

subsequent fermentation can take place. Incomplete separation of lignin and cellulose, as

well as the presence of by-products during fermentation could inhibit the bio-ethanol

formation (Cheng, 2010).

2.2 Biomass Properties

Calorific value is the measure of heat energy released when a specific quantity of

fuel undergoes complete combustion. This analysis is often conducted in a bomb

calorimeter under constant-volume condition, and expressed as (J/kg). Energy content in

biomass is influenced by factors such as moisture content, lignin content, oxygen: carbon

ratio and hydrogen: carbon ratio. In general, the heating value of biomass is comparable to

low rank coal at around 10-20 MJ/kg due to its high oxygen and ash content (Saidur et al.,

2011). Properties of biomass in comparison to bituminous coal are tabulated in Table 2-2.

Literature Review

17

Table 2-2: Biomass properties. Published with permission from (Jenkins et al., 1998)

Coala Wheat straw Rice straw Switch-grass

Proximate analysis (% dry fuel)

Fixed carbon 77 17.71 15.86 14.34

Volatile matter 18.49 75.27 65.47 76.69

Ash 4.51 7.02 18.67 8.97

Total 100 100 100 100

Ultimate analysis (% dry fuel)

Carbon 87.52 44.92 38.24 46.68

Hydrogen 4.26 5.46 5.2 5.82

Oxygen (diff.) 1.55 41.77 36.26 37.38

Nitrogen 1.25 0.44 0.87 0.77

Sulphur 0.75 0.16 0.18 0.19

Chlorine 0.16 0.23 0.58 0.19

Ash 4.51 7.02 18.67 8.97

Total 100 100 100 100

Elemental composition of ash (%)

SiO2 37.24 55.32 74.67 65.18

Al2O3 23.73 1.88 1.04 4.51

TiO2 1.12 0.08 0.09 0.24

Fe2O3 16.83 0.73 0.85 2.03

CaO 7.53 6.14 3.01 5.6

MgO 2.36 1.06 1.75 3

Na2O 0.81 1.71 0.96 0.58

K2O 1.81 25.6 12.3 11.6

SO3 6.67 4.4 1.24 0.44

P2O5 0.1 1.26 1.41 4.5

CO2/other

Total 98.2 100 100 100

Undetermined 1.8 1.82 2.68 2.32

Higher heating value (constant volume)

MJ/kg 35.01 17.94 15.09 18.06 a :

Low volatile bituminous

Proximate analysis indicates the biomass composition with regards to moisture

content, volatiles, fixed carbon, and ash content. It provides an overview on the thermal

behaviour of biomass with respect to devolatilisation temperature, ignition temperature,

and rate of decomposition. The composition is dependent on temperature, heating rate,

nature of biomass, and the presence inorganic species (McKendry, 2002) (Jameel et al.,

2010).

Literature Review

18

Biomass contains relatively higher moisture and volatiles, and lower fixed carbon

compared to coal. Devolatilisation of biomass yields more oxygen functional groups,

e.g.: hydroxyl, carbonyl (-COOH, -OH) which are reactive at low temperature and gives

biomass its low heating value (Vassilev et al., 2010). This results in a greater reduction in

mass (up to 90 wt.%) of initial biomass compared to less than 10 wt.% in coal at this stage

(Jenkins et al., 1998). For instance, volatile matter constitutes around 75 wt.% of biomass

composition compared to less than 20 wt.% in coal. Fixed carbon in biomass ranges

between 14-20 wt.% relative to 77 wt.% found in coal.

The combustion of fixed carbon leaves behind residual ash. These ashes can be

grouped into two categories; inherent ash and entrained ash. The former is naturally

occurring and intimately distributed throughout the fuel while the latter is normally

entrained during biomass processing steps. Alkali metals, alkaline-earth metals, and salts

are commonly found in biomass. These include potassium, sodium, magnesium, silica, and

phosphorus. The presence of alkali metals that are inherently volatile, most notably K and

Na could be problematic. The reaction between alkali metal and silica produce alkali

silicates which have low ash melting point, T

Literature Review

19

The composition of organic elements in biomass is determined from ultimate

analysis. Elements such as carbon (C), hydrogen (H), and oxygen (O) constitute the major

composition of biomass, along with a smaller fraction of nitrogen (N) and sulphur (S). This

helps in the determination of biomass heating values and environmental impact associated

with the release of sulphur, nitrogen, and chlorine during the combustion process (Saidur

et al., 2011). In general, biomass has higher H and O content, but relatively lower C, N,

and S compared to coal. It is a highly oxygenated compound with around 30-40 wt.%

oxygen, compared to just below 2 wt.% in coal. Its high H: C and O: C ratio explains its

low heating value. This is due to lower energy contained in C-O and C-H bonds, than in C-

C bonds (McKendry, 2002). However, Van Loo and Koppejan (Van Loo & Koppejan,

2008) argued that C and H contribute positively to heating value, since both are oxidised

during combustion to form CO2 and H2O, respectively. Biomass is ideal for co-combustion

with coal, despite its relatively lower heating value. Lower S and N content means reduced

pollutant from high temperature combustion.

Figure 2-2: Bulk density of biomass relative to coal (Clarke & Preto, 2011)

0

100

200

300

400

500

600

700

800

Loose straw Baled straw Sawdust Hardwood

chips

Pelletised

straw

Lignite

Bu

lk d

ensi

ty (

kg

/m3)

Samples

Literature Review

20

Bulk density is the weight of a known volume of biomass, expressed as (kg/m3).

Bulk density of biomass is presented in Figure 2-2. In general, biomass has low density

and low energy content relative to coal. Hence, a bigger volume of biomass is needed to

achieve the equivalent energy produced by coal. These factors are making it less attractive

for large-scale energy generation and pose challenges in terms of feeding system,

transportation, storage, and the choice of conversion technologies (Van Loo & Koppejan,

2008). For biomass to be cost-competitive, a densification process is suggested to increase

its energy density. Pelletisation, briquetting, and baling are among common densification

steps. For example, pelletisation of loose straw increases its density from 20 kg/m3 to 635

kg/m3.

2.3 Pyrolysis

2.3.1 Mechanism

The breaking of molecular bonds within biomass structure under endothermic

reactions releases hot vapours or aerosols. Rapid quenching of vapours produces bio oil or

tar, while non-condensable vapours form permanent gases. These gases are mainly made

up of hydrogen, carbon dioxide, carbon monoxide, methane, and a smaller fraction of light

hydrocarbons, e.g.: ethylene and ethane (Basu, 2010c). It is widely acknowledged that

pyrolysis proceeds in two stages; primary pyrolysis and secondary pyrolysis depending on

local temperature as illustrated in Figure 2-3. This corresponds to the decomposition

temperature of hemicellulose, cellulose, and lignin which governed the heat and mass

transfer processes. Devolatilisation rate is influenced by several factors such as heating

rate, reaction temperature, moisture content, and the presence of catalyst.

Literature Review

21

The process begins with biomass drying at a temperature around 100°C. Free or

loosely bound water is driven off, allowing heat to penetrate further into biomass. Further

increase in temperature initiates chemical reactions, leading to the breakdown of organic

functional groups within biomass. Broadly speaking, pyrolysis can be grouped into four

reactions; random bond scission, depolymerisation, carbonisation, and side reactions

(Babu, 2008). The scission of oxygen functional groups; i.e.: OH, C=O at low temperature

leads to the formation of free radicals, carbonyl and carboxyl group. This is indicated by a

greater evolution of CO, CO2, and water vapour at low temperature.

Figure 2-3: Pyrolysis stages. Published with permission from (Neves et al., 2011)

Main pyrolysis begins at a temperature around 350°C, indicated by the release of

aerosols. Depolymerisation reaction removes the monomer units from biomass, which

participate in the formation of free radicals and chain reactions. This influences the gas

composition. For example, the breaking of C- H alkyl results in the formation of CH4 at

low temperature, between 300-500°C. Dehydrogenation of C-H bond at T>450°C

contributes to increased H2 production. This is due to the breakdown of stable aromatic

Literature Review

22

rings which require higher dissociation energy. For instance, bond dissociation energy

required to remove H from CH3 is 435.2 kJ/mol.

Hot volatiles and char from primary pyrolysis are vulnerable to the risk of

exothermic secondary reactions. These reactions could occur in the vapour phase or

between vapour-solid phases. These include the thermal cracking of volatiles,

re-polymerisation of lighter organic compounds in bio oil, char gasification, and shift

reaction. Long vapour residence time in a hot reactor could induce thermal cracking of

heavy hydrocarbons into lighter hydrocarbons and permanent gases. Apart from high

temperature, the presence of char has a catalytic effect, which causes tar cracking into

gases. Fast removal of char from a pyrolysis reactor although recommended, is inefficient.

Increased H2 and CO at higher temperature also imply the extent of secondary cracking of

hydrocarbon gases into lighter, permanent gases (Dai et al., 2000).

These volatiles escape through pore openings that enlarge with an increase in

temperature. This is observed by an increase in char porosity. Carbonisation is the

polymerisation of radicals, achieved through elimination of side chains to form stable

chemical structures. These concurrent reactions include aromatisation, and cyclisation of

alkyl chains through dehydrogenation. Volatiles released during pyrolysis contributes to

the formation of carbonaceous char, of high CV~ 20-30 MJ/kg (Babu, 2008)

2.3.2 Cellulose Decomposition

Several reaction pathways have been proposed to achieve better understanding of

cellulose decomposition. The first reaction model was the Broido model as illustrated in

Figure 2-4. In this model, cellulose is initially converted into ‘active’ cellulose at elevated

temperature. The active cellulose then reacts via two parallel pathways; forming volatile

Literature Review

23

tars and solid intermediate (A). The latter undergo consecutive reactions to form char B

and C, with accompanying volatiles.

Figure 2-4: Broido model of cellulose decomposition (Varhegyi & Jakab, 1994)

This is later simplified by Shafizadeh by omitting the reactions leading to the

formation of char B and C. The Broido-Shafizadeh model proposed that cellulose

decomposes into a combination of (char + gases) and condensable volatiles via two

competing reactions, as shown in Figure 2-5. Several temperature-dependent reactions

occur during cellulose decomposition. These reactions are dehydration, depolymerisation,

and fragmentation. Initially, a low degree of polymerisation takes place to convert

‘inactive’ cellulose into active cellulose. This initiation step requires a high activation

energy (EA) of 242.7kJ/mol, yet only a 3-6% mass loss is observed during this period

(White et al., 2011)

Literature Review

24

Figure 2-5: Broido-Shafizadeh model of cellulose decomposition. Published with permission from

(Basu, 2010c)

This is followed by two competing reactions; dehydration and depolymerisation.

Cellulose dehydration is followed by Reaction II. Decarboxylation and carbonisation

reactions are responsible for the formation of free radicals, water soluble acids, anhydrides,

water vapour, CO2, CO, and char. This reaction proceeds at a much lower temperature

around 200-280°C and slow heating rate.

Meanwhile, cellulose depolymerisation is favoured at high temperature around

280-340°C and fast heating rate. The breakage of glycosidic bonds (Reaction III) yield

condensable volatiles. This reaction possesses higher activation energy (EA) at 198 kJ/mol

than dehydration at 153 kJ/mol. It is also responsible for the production of anhydrosugar;

e.g.: levoglucosan, oligosaccharides, hydroxyl-acetaldehydes, acids, and alcohol. Under

very fast heating rate; e.g.: flash pyrolysis where T>500°C, direct conversion of cellulose

to condensable volatiles and permanent gases through fragmentation may occur. This is

achieved via fission, disproportionation, dehydration, and decarboxylation reactions with

no char formation (White et al., 2011).

Literature Review

25

Secondary reactions (Reaction IV) are also included in this model, which proceed

at a much lower EA = 107.5 kJ/mol. This means the thermally unstable ‘intermediates’ are

vulnerable to react further to non-condensable gases, char, and tar as mentioned previously

in Section 2.3.1. Thus, it is important to quickly extract the condensable vapours from the

hot reactor. This can be achieved by increasing the flow rate of purge gas and having a

small reactor.

2.3.3 Bio Oil Properties

Bio oil is a homogeneous mixture of aqueous and organic phase, and is usually

dark-brownish in colour. It is a product of rapid and simultaneous depolymerisation and

fragmentation of lignocellulosic components during thermal decomposition of biomass. It

mainly consists of water-soluble compounds such as acids, ketones, phenols, furfurals, and

aromatic hydrocarbons (Bridgwater, 2012). Biomass dehydration results in bio oil with

high water content, at around 15-30 wt.%., with low energy density. This poses several

problems in downstream applications.

High water content reduces the vaporisation rate of oil droplets and increases

ignition delay time in a combustion engine. It risks lowering the combustion temperature

and induces flame instability problems. The presence of char residues or alkaline metals,

i.e.: potassium in bio oil, could catalyse the ageing of bio oil during storage. It can be

described as re-polymerisation of lighter organic compounds to form heavier tar, indicated

by phase separation of bio oil into aqueous phase and lignin-rich phase. This reaction

produces more water as by-product. High water content (>30 wt.%) also has been found to

induce a phase separation problem (Lehto et al., 2014).

Literature Review

26

Bio oil is highly oxygenated. Analysis shows that it consists around 35-40 wt.%

oxygen, with low H: C ratio. The latter could induce instability problem during mixing with

conventional oil. The presence of weak acids, e.g.: acetic acid in bio oil poses challenge

during storage and risk of corrosion in the boiler. Typical pH range of untreated bio oil is

found to be between 2.5 to 3. Unlike crude oil, distillation of bio oil is practically impossible

since it is made up of water-soluble compounds that are inherently volatile. Fractional

distillation or heating above 100°C could cause losses of valuable organic compounds such

as xylene. Moreover, continuous heating at high temperature induces thermal cracking of

tar into char, at around 45 wt.% of original bio oil. Heating value of bio oil is found to be

half of conventional hydrocarbon oil at around 20 MJ/kg.

2.3.4 Tar Composition

Tar can be grouped into five different classes; depending on its molecular weight

and solubility in water as shown in Table 2-3. Very heavy tar compounds are represented

by Tar Class 1. This is followed by heterocyclic aromatic compounds, with -O and -N

attached to the aromatic rings. Phenol C6H6O and pyridine C5H5N are representative of tar

class 2. Mono-aromatic hydrocarbons such as benzene, toluene, and styrene with single

ring can be classified as tar class 3. Meanwhile, light PAH compounds with 2-3 benzene

rings such as indene, naphthalene and fluorine are representative of tar class 4. Compounds

with more than three benzene rings such as pyrene and chrysene can be classified as heavy

PAH compounds.

Temperature exerts a strong influence on tar formation as depicted in Figure 2-6.

The presence of mixed oxygenates, e.g.: levoglucosan, acetic acid, furfural and catechol

are observed during low temperature pyrolysis, T~400°C. Lignin decomposition at high

temperature, T>500°C contributes to increased formation of phenols and mono-aromatic

Literature Review

27

hydrocarbons. However, a further increase in temperature up to 900°C results in the

formation of heavy PAH compounds. This is believed to be caused by two reactions; the

condensation of light aromatic compounds into heavier tar and the cracking of very heavy

tar (tar class 1) into tar class 4 and 5.

Table 2-3: List of tar compounds. Published with permission from (Li & Suzuki, 2009)

Tar

class Class name Property Representative compounds

1 GC-undetectable Very heavy tars, cannot be

detected by GC

Determined by subtracting the GC-

detectable tar fraction from the total

gravimetric tar

2 Heterocyclic aromatics

Tars containing hetero

atoms; highly water

soluble compounds

Pyridine, phenol, cresols, quinoline,

isoquinoline, dibenzophenol

3 Light aromatic (1 ring)

Usually light hydrocarbons

with single ring; do not

pose a problem regarding

condensability and

solubility

Toluene, ethylbenzene, xylenes,

styrene

4 Light PAH compounds

(2–3 rings)

2 and 3 rings compounds;

condense at low

temperature even at very

low concentration

Indene, naphthalene,

methylnaphthalene, biphenyl,

acenaphthalene, fluorene,

phenanthrene, anthracene

5 Heavy PAH compounds

(4–7 rings)

Larger than 3-ring; these

components condense at

high-temperatures at low

concentrations

Fluoranthene, pyrene, chrysene

Figure 2-6: Tar formation scheme. Published with permission from (Li & Suzuki, 2009)

Literature Review

28

2.3.5 Pyrolysis Types

As pointed out above; heating rate, operating temperature, hot vapour residence

time, and feedstock properties are among factors that influence pyrolysis product yield. The

latter incudes moisture content, particle size, and the presence of catalyst which influence

the rate of devolatilisation and heat transfer efficiency. Depending on temperature and

heating rate, pyrolysis can be grouped into slow pyrolysis, fast pyrolysis, and flash

pyrolysis as shown in Table 2-4. Heating rate varies in the range of 0.1-1000 °C/s, with no

clear boundary between slow and fast pyrolysis. However, it is accepted that heating rate

>5°C/s is set as a common threshold for fast pyrolysis (Neves et al., 2011)

Slow pyrolysis is conducted at moderate temperature (~600°C) and residence time

~4s to produce char, oil, and gas. Meanwhile, fast and flash pyrolysis are optimised for

maximum bio oil yield. Flash pyrolysis is characterised by very high heating rate

>1000°C/s and short vapour residence time of less than 1s, capable of producing ~75% oil

yield. Pulverised biomass with high surface area is usually employed in flash pyrolysis to

ensure rapid heating.

Table 2-4: Types of pyrolysis. Adapted from (Jahirul et al., 2012)

Pyrolysis type Heating rate Temperature (°C) Residence time Major products

Slow pyrolysis Low ~600 Long, 4s Char, oil, gas

Fast pyrolysis High ~400-650, 650-900 Short, 1s Oil ~ 60-75%

Flash pyrolysis High 450-1000 Very short,

Literature Review

29

heating applications and chemical synthesis. It has a frequency range between 300 MHz to

300 GHz, although 0.915-2.45 GHz is used as standard in industrial and domestic

microwave oven applications. This is done to avoid interference with ultra-high frequency

used in radar and telecommunication applications (Fernandez et al., 2011).

Figure 2-7: Electromagnetic spectrum. Published with permission from (Motasemi & Afzal,

2013)

Microwaves are a type of transverse waves; where electric field (E) and magnetic

field (H) are oscillating at a right angle or perpendicular to each other. This is illustrated in

Figure 2-8. Microwave heating is the result of microwave energy absorption by dielectric

materials. In general, materials can be grouped into three different categories;

i.e.: conductor, insulator, and absorber based on their dielectric behaviour within an

electromagnetic field.

Oven cavity is an example of perfect electrical conductor, where microwaves are

impenetrable and reflected throughout. On the other hand, insulator is a transparent material

which permits microwaves to travel through it without incurring any dielectric loss.

Materials with high penetration depth are commonly used as microwave insulators. These

include quartz and Teflon, with penetration depth of 160 m and 92 m, respectively.

Literature Review

30

Meanwhile, materials that absorb microwave energy such as wood, water, graphite are

commonly known as microwave absorber or dielectric materials.

Figure 2-8: Electric (E) and magnetic (H) field components in microwave. Published with

permission from (Motasemi & Afzal, 2013)

These materials have a higher tendency to be polarised by an oscillating electric

field. The likelihood of molecules to follow the alternating field results in the heat

generation through molecular friction and dielectric loss. This is otherwise known as

dielectric polarisation. Ionic liquids such as water and acid have a higher chance of being

polarised, compared to non-polar substances such as beeswax. One important aspect of

dielectric polarisation is response time (RT) of dipoles relative to microwave frequency. It

is preferable for RT to match or slightly lag behind the changing electric field, for molecular

friction to occur. This condition is met at microwave frequency of 2.45 GHz, which allows

the dipoles to align in the field, but not to follow alternating field precisely (Yin, 2012).

Heat generated during this process is known as dielectric heating or commonly known as

microwave heating.

Literature Review

31

Figure 2-9: Complex permittivity of dielectric material

The ability of dielectric material to absorb microwave radiation and converting it

into heat is characterised by complex permittivity (ε). This refers to the real and imaginary

parts in the complex plane, such that ε = ε’-jε’’, as illustrated in Figure 2-9. Dielectric

constant (ε’) is the measure of material to conduct and store electrical energy in an electric

field. Meanwhile, dielectric loss (ε’’) is a term referring to the ability of material to dissipate

this energy into heat. Loss factor (α) defines the ability of material to convert

electromagnetic radiation into heat at a specified frequency and temperature.

2.4.1 Microwave Modes

In general, microwave ovens can be classified into single-mode and multi-mode

cavities. An example of a multi-mode cavity is domestic microwave oven. Multiple

radiation is reflected within the enclosed metallic cavity, creating a field with simultaneous

resonant modes. This generates multiple hot and cold spots within the cavity, which

influences the heating uniformity of the sample load. Heat distribution can be improved by

introducing a mode stirrer and rotating table into the oven. For instance, the presence of a

mode stirrer disturbs the standing wave pattern inside the oven cavity. This continuously

moves the location of hot and cold spots throughout the cavity.

Literature Review

32

Meanwhile, a rotating table is an example of travelling load. Uniform heating is

achieved as the load travels across the location of maximum and minimum electromagnetic

field. In contrast, only a single mode of radiation is generated in single mode cavity. This

creates a well-defined electromagnetic field, with known region of high and low energy

intensity. This allows a sample to be placed in a particular hot spot for maximum energy

conversion. The nature of focused radiation in a single mode cavity is particularly useful

during high temperature processing of low-loss materials; e.g.: ceramic (Acierno et al.,

2004)

2.4.2 Conventional Heating versus Microwave Heating

Differences in heating methods are illustrated in Figure 2-10. Pre-heated air, radiant

furnace, and fluidised bed are typical examples of external heat sources for particle drying

in conventional heating. This process however is often slow and inefficient. A non-linear

temperature gradient is observed in this type of heating. The particle surface records a

higher temperature than the core, as heat propagates from outer regions inwards. Heat

transfer efficiency is often determined by thermal properties such as thermal conductivity,

density, specific heat and also convection heat transfer coefficient (Motasemi & Afzal,

2013).

In contrast, the conversion of electromagnetic energy into heat energy results in

volumetric heating throughout the particle during microwave heating. Particle core records

a higher temperature than the surface. The internal heat generation causes moisture

evaporation and initiates the subsequent chemical reactions.

Literature Review

33

Figure 2-10: Difference in heating process. Published with permission from (Motasemi & Afzal,

2013)

Microwave heating has proven to be beneficial in the context of pyrolysis. This can

be explained by the interaction of hot volatiles as they progress from core towards the

particle surface. In the case of conventional heating, these volatiles have to travel through

the hotter region of the particle, leading to thermal cracking into gases. However, these

volatiles progress through a colder region of the particle during microwave heating;

reducing risk of secondary reactions (Miura et al., 2004).

2.5 Previous Work on Microwave Pyrolysis of Biomass

Microwave pyrolysis has gained interest in recent years. It is seen as an alternative

which could lift the strict requirements associated with conventional pyrolysis. These

include particle size, moisture, poor heat transfer, and high furnace temperature which

favour secondary reactions. It utilises a similar setup to conventional pyrolysis. An inert

environment is maintained by a continuous flow of non-reactive gas; e.g.: N2, and bio oil

recovery is achieved through rapid quenching of volatiles. The only difference lies in their

heating processes.

Literature Review

34

Investigation into microwave pyrolysis has covered various biomass materials such

as rice straw (Huang et al., 2008), coffee hulls (Domínguez et al., 2007), waste automotive

oil (Lam et al., 2012), oil palm biomass (Salema & Ani, 2011), sewage sludge (Domínguez

et al., 2005) (Tian et al., 2011), wheat straw (Krieger-Brockett, 1994)(Zhao et al., 2012),

waste paper (Popescu et al., 2008), corn stover (Wan et al., 2009) (Lei et al., 2009) and

microalgae (Du et al., 2011)(Beneroso et al., 2013). This strongly suggests that microwave

pyrolysis can be applied to a wider range of biomass regardless of feedstock size and water

content. Microwave pyrolysis eliminates cost associated with biomass pre-treatment such

as drying and grinding.

Among parameters studied in the microwave pyrolysis of biomass include particle

size, microwave power, reaction temperature, and addition of catalyst and microwave

absorber. The effect of particle size is found to be insignificant during microwave pyrolysis

than in conventional pyrolysis, as demonstrated by (Lei et al., 2009). Ground stover

particles with d=0.5-4 mm are found to be similarly pyrolysed. Miura et al. (Miura et al.,

2004) have successfully demonstrated the microwave pyrolysis of wood blocks with

diameter ranging between 0.06 to 0.3 m. A cross section of a wood block was examined

after three minutes into the pyrolysis process, as shown in Figure 2-11. It was revealed that

carbonisation starts from the centre of the wood, which confirmed the nature of volumetric

heating.

Literature Review

35

Figure 2-11: Cross section of wood block, ϕ= 80 mm. Published with permission from (Miura et

al., 2004)

In addition to that, microwave heating is found to enhance or speed up the chemical

reactions during pyrolysis. These reactions proceed at a much lower temperature during

microwave pyrolysis, compared to conventional pyrolysis. This has been demonstrated by

Budarin et al. (Budarin et al., 2009) with low temperature activation of wheat straw, where

bio oil is produced at a claimed temperature below 180°C, compared to 350°C during

conventional processing. Rapid microwave heating also means reduced reaction time. This

is achieved in a matter of just several minutes, compared to ~40 minutes in a standard fixed-

bed pyrolysis reactor. The latter would require the reactor to be heated up until it has

reached desired temperature before biomass introduction. Better efficiency has resulted in

total energy saving from microwave pyrolysis, as indicated by (Lam et al., 2012; Zhao et

al., 2011).

The effect of microwave power has been investigated by (Hu et al., 2012; Tian et

al., 2011; Du et al., 2010; Yu et al., 2007). 300 W is found to be the minimum power to

initiate pyrolysis process, with low bio oil yield obtained at P ≤ 480 W. An increase in

microwave power has resulted in increased production of gas and bio oil at the expense of

Literature Review

36

reduced char yield. This is due to the sample being subjected to a higher heating rate.

During microwave pyrolysis of Chlorella vulgaris, sample temperature rose from 200 to

800°C, when microwave power was increased from 750 to 2250 W. Maximum gas yield of

52.37 wt.% is also recorded at maximum wattage.

The effect of reaction temperature on product yield and distribution was

investigated. Slow devolatilisation at low temperature (T< 400°C) resulted in low oil and

gas yield. An increase in temperature caused more condensable volatiles to be released,

until it reaches certain temperature.